Physical Injuries and Hair Corticosterone Concentration in Rabbit Kits from Single- and Group-Housed Does Kept on a Commercial Farm

Simple Summary On commercial farms, adult female rabbits (does) are usually kept in single cages without direct contact with conspecifics other than their kits. However, wild rabbits live together in groups with several adults and their kits, and single housing of adult domestic rabbits may affect both, their own welfare and the welfare of their kits. In this study, we focused on the welfare of kits and compared three different housing systems for does and their kits: single housing of does with their respective kits; pairs of does with their kits; and groups of three does with their kits. The kits’ health and stress level were analyzed. The number of kits with skin lesions increased with increasing group size. However, when does were pair-housed, kits seemed to have fewer disease symptoms compared to the other treatments. With regard to the stress level, there was no clear result; however, kits in groups of three does did tend to have higher stress hormones in hair. We conclude that pair housing for does could be an alternative to single housing when considering the welfare of their kits. Further studies are needed to evaluate both, the effects of pair housing on kits and does. Abstract In intense breeding programs, rabbits are exposed to numerous stress factors which could affect their welfare and health. It has been suggested that group housing is more comfortable for does and similar to the living conditions of European wild rabbits. In this study, we compared three different housing systems—single housing; housing in pairs; and housing in groups of three does—to test whether there is a measurable impact on skin lesions, health, and hair corticosterone concentration (HCC) of their kits. The number of kits with lesions increased with the number of does kept together. The probability of kits getting injured was higher in groups of three does than in pens of single-housed does (p = 0.041). When does were pair-housed, kits seemed to have fewer disease symptoms compared to the other treatments. Concerning HCC of kits, there was no significant difference between the housing systems (p > 0.05). The mean HCC of kits was 2.94 pg/mg, while pair housing had the lowest HCC (2.59 pg/mg). This study focused on the welfare of kits from group-housed does. From this perspective, pair housing of does may be appropriate to allow social interaction between does without unduly affecting the welfare of their kits.


Introduction
Although rabbits live in social groups in the wild, adult domestic rabbits are commonly individually housed in intensive farming systems, even though it is known that singlehoused rabbits often perform abnormal behaviors (digging, floor chewing, bar biting), which is an indicator of compromised welfare [1][2][3]. The gregarious nature of wild rabbits

Animals, Housing, and Handling
The present study was carried out on a commercial rabbit farm in Lower Saxony, Germany, where approximately 600 rabbits for breeding purposes (HYPLUS PS 19) and their offspring (HYPLUS PS 19x PS 59, HYPHARM S.A.S., Sèvremoine, France) were kept in a semi-intensive reproduction system with a 42 d reproductive rhythm and artificial insemination at day 11 postpartum (pp).
For this study, a housing system for does with 36 individual units was used. These individual units could be joined together to form group pens for a variable number of does. For this purpose, the units were connected by doors (25 cm × 22 cm). The size of each individual unit was about 80 cm × 80 cm. The units had wire mesh walls and were open at the top. The plastic slatted floor of each unit had 11-mm-wide slots and slats. On the back wall of each unit, an elevated platform (58 cm × 52 cm) was installed. The elevated platform was not fully slatted, since part of it (40 cm × 52 cm) was covered by a plastic mat. The platform was inclined by 9% to drain off urine. Environmental enrichment was offered in each individual unit: a plastic tube (placed above the platform), a chain hanging from the platform, and gnawing material (a piece of wood attached to a chain hanging from the platform, and a second piece of wood as well as a cotton rope, both attached to the wall). The morning after the does had entered the housing system, nest boxes (25 cm × 39 cm × 32 cm) attached to the outside of each individual unit and filled with wood shavings were opened up for nest building. Figure 1 depicts the housing system, which has already been described by Bill et al. [25]. The animals were fed a diet for lactating does (Fok Lapin, Victoria Mengvoeders B.V., Veghel, The Netherlands) ad libitum; additionally chopped hay was offered. Furthermore, each unit was equipped with two nipple drinkers whereby water was constantly available. Manure was collected in a slurry store. During the study period, the mean temperature in the vacuum ventilated building was 19.5 ± 2.4 • C, and the mean humidity was 77.8 ± 9.8%. The housing system was lit by LED light strips that were switched on from 05:30 h to 19:00 h with dimming sequences of half an hour each. In addition, there were windows allowing daylight to enter the stable [25].
The present study was carried out on a commercial rabbit farm in Lower Sa Germany, where approximately 600 rabbits for breeding purposes (HYPLUS PS 1 their offspring (HYPLUS PS 19x PS 59, HYPHARM S.A.S., Sèvremoine, France) wer in a semi-intensive reproduction system with a 42 d reproductive rhythm and ar insemination at day 11 postpartum (pp).
For this study, a housing system for does with 36 individual units was used. individual units could be joined together to form group pens for a variable num does. For this purpose, the units were connected by doors (25 cm × 22 cm). The each individual unit was about 80 cm × 80 cm. The units had wire mesh walls and open at the top. The plastic slatted floor of each unit had 11-mm-wide slots and sla the back wall of each unit, an elevated platform (58 cm × 52 cm) was installed. Th vated platform was not fully slatted, since part of it (40 cm × 52 cm) was covere plastic mat. The platform was inclined by 9% to drain off urine. Environmental e ment was offered in each individual unit: a plastic tube (placed above the platfo chain hanging from the platform, and gnawing material (a piece of wood attache chain hanging from the platform, and a second piece of wood as well as a cotton both attached to the wall). The morning after the does had entered the housing s nest boxes (25 cm × 39 cm × 32 cm) attached to the outside of each individual un filled with wood shavings were opened up for nest building. Figure 1 depicts the ho system, which has already been described by Bill et al. [25]. The animals were fed for lactating does (Fok Lapin, Victoria Mengvoeders B.V., Veghel, The Netherlan libitum; additionally chopped hay was offered. Furthermore, each unit was equ with two nipple drinkers whereby water was constantly available. Manure was co in a slurry store. During the study period, the mean temperature in the vacuum lated building was 19.5 ± 2.4 °C, and the mean humidity was 77.8 ± 9.8%. The ho system was lit by LED light strips that were switched on from 05:30 h to 19:00 h dimming sequences of half an hour each. In addition, there were windows all daylight to enter the stable [25].  (5), and doors (6) for joining ind units to a group pen (adapted with permission from Bill et al. [25]. 2020, Elsevier). Does were moved to three different housing treatments (single housing, in p groups of three does) one week before parturition. Two to three days before kin doors between single units were closed and all does were single-housed. After 18 the doors were opened again and the does returned to their groups, now togethe  (5), and doors (6) for joining individual units to a group pen (adapted with permission from Bill et al. [25]. 2020, Elsevier). Does were moved to three different housing treatments (single housing, in pairs or groups of three does) one week before parturition. Two to three days before kindling, doors between single units were closed and all does were single-housed. After 18 days, the doors were opened again and the does returned to their groups, now together with their kits (postpartal mixing). Until the day of insemination, controlled suckling took place once a day by opening the nest boxes manually. Afterwards, nest boxes were constantly open to enable free suckling. At day 21 pp, all nest boxes were closed to fortify kits to eat solid food [25]. In total, 60 kits were selected for the study, 20 per type of housing ( Table 1). Half of the kits were male, the other half female.

Lesion Scoring and Health Status
At day 30 pp, i.e., the day of weaning, all kits were weighed individually and their health status was scored as either 0 = adspectorally healthy or 1 = with disease symptoms (additionally all symptoms were recorded separately). The occurrence of injuries was scored for each individual kit according to the following scheme: 0 = no injuries; 1 = slightly and superficially injured; 2 = deep injuries; and 3 = partial loss of body parts (e.g., ears).

Hair Samples
At day 30 pp, hair samples for corticosterone measurement were collected. Using commercially available electric clippers for small animals, a symmetric area of 2 cm × 2 cm was shaved laterally on the hind leg as close as possible to the skin. The hair was stored in airtight plastic bags under light-protected and dry conditions at room temperature to avoid a possible washout effect by UV radiation [15]. The hair samples were sent to the Dresden LabService GmbH, Dresden, Germany for analysis. One sample per kit was collected and analyzed.

Corticosterone Extraction
The impact of contamination by sweat, sebum, or excreta may be minimal if samples are washed [15,19,20]. The hair washing and steroid extraction procedures were based on the protocol described in Stalder et al. [26] (study II) with modifications made by Gao et al. [27] to allow analysis by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS): Hair strands were washed by shaking them in 2.5 mL isopropanol for 3 min at room temperature. They were then allowed to dry under a fume hood for at least 12 h. A total of 10 mg of whole, non-pulverized hair was carefully weighed and transferred into a 2 mL tube (Eppendorf AG, Hamburg, Germany). After this, 50 µL internal standard and 1.8 mL methanol were added, and the hair was incubated for 18 h at room temperature for steroid extraction. Samples were spun in a centrifuge at 10,000 rpm for 2 min and 1 mL of the clear supernatant was transferred into a new 2 mL tube. The alcohol was evaporated at 65 • C under a constant stream of nitrogen until the samples were completely dried (duration: approximately 20 min). The dry residue was re-suspended using 250 µL distilled water, 200 µL of which was used for LC-MS/MS analysis [27].

Corticosterone Concentration by LC-MS/MS
The HCC was analyzed using LC-MS/MS, which eliminates cross-reactivity, since each steroid has a unique molecular mass-to-charge ratio, thus ensuring high specificity [28]. The lower detection limit of this protocol is 0.3 pg/mg for corticosterone. All samples were analyzed in one charge. The intra-assay coefficient of variation (variation within plates) of this assay was below 9.3%.

Statistical Analysis
Statistical analyses were performed using the statistical software R [29]. The level of significance was set at p < 0.05. Data were checked for normality using histograms. The measured corticosterone values were logarithmized to approach a normal distribution. Since no kits had a lesion score of 2 and only two kits a score of 3, lesion scores of 1 to 3 were combined for the statistical models, resulting in a binary lesion variable, with the feature values "yes" = any type of lesions, or "no" = no lesions.

Lesions
To test whether the group size of the does had an impact on the lesions of the kits, at first, a logistic regression model [30] was used with lesion as dependent variable and group size as independent variable. The Akaike information criterion (AIC) [30] was calculated to assess the model quality. Afterwards, further independent variables like sex, weight, health status, and HCC were added individually to the model to see whether they decreased the AIC of the model. Finally, an F-test was conducted for the model with the lowest AIC, with group size and weight as independent variables to test which of these independent variables had a significant impact on the lesion variable. In case of significance, pairwise comparisons using Tukey's method [31] were additionally conducted using the R package "emmeans" [32].

Hair Corticosterone Concentration
To test whether different factors had an impact on the HCC of the kits, a linear regression model [30] was used with logarithmized HCC as the dependent variable and group size, sex, health status, and lesion as independent variables. Again, an F-test was conducted. Since no significant results were obvious in the F-test, no pairwise comparisons were carried out.

Body Weight, Lesions, and Health Status
The mean body weight of all kits at weaning was 765.15 g ± 123.08 g. Kits from groups of three does had the lowest body weight at weaning, with a mean of 711.55 g ± 116.44 g, followed by kits of single-housed does (729.35 g ± 112.55 g). Kits of pair-housed does had the highest mean body weight at weaning (854.55 g ± 84.10 g). A total of 19 of 60 kits (five males and 14 females) had disease symptoms, with the majority of kits suffering from rhinitis (13 kits, i.e., 68.4% of the diseased kits). Other disease symptoms of kits were conjunctivitis in one or both eyes, mild pollution of the genitals, and central nervous symptoms. Kits of pair-housed does had the lowest number of symptoms (20%), followed by kits from groups of three does (35%), and kits from single-housed does (40%). A total of 29 kits (12 males and 17 females) had skin lesions. The number of kits with lesions increased with the number of does per group, and therefore also with the number of kits kept together (kits from single-housed does = 25%, kits from pair-housed does = 55%, kits from groups of three does = 65%).
The results of the logistic regression model are shown in Tables 2-4. The results of the F-test showed significant differences regarding the risk of injury between at least two group sizes (p = 0.026), but not for body weight (p > 0.05) ( Table 3). Compared to kits from single-housed does, the odds of getting injured increased when kits were kept in cages with pair-housed does (p = 0.056) and they increased significantly when kits were raised Animals 2023, 13, 196 6 of 12 in cages housing three does (p = 0.041). There was no significant difference between the housing treatments pairs and groups of three (p > 0.05) ( Table 4).

Hair Corticosterone Concentration
The mean HCC of all kits was 2.94 pg/mg ± 1.06 pg/mg (minimum: 1.29 pg/mg; maximum: 7.48 pg/mg). The results of the linear regression model are shown in Tables 5  and 6. There was no significant impact of group size, sex, health status, and lesions on the HCC of kits (Table 6). However, it was descriptively shown that kits originating from groups of three does tended to have higher HCC values than kits from single-housed and pair-housed does ( Figure 2). This was also reflected in the estimates ( Table 5). Kits of pair-housed does had the lowest HCC, with a mean of 2.59 pg/mg, followed by kits of single-housed does with a mean of 2.91 pg/mg. The highest mean HCC was found for kits originating from groups of three does (3.34 pg/mg). Female kits had slightly higher HCC values than male kits (Figure 3).

Discussion
In this pilot study, HCC of rabbit kits were presented as a possible new marker for the welfare of young animals from birth to weaning. Based on HCC and skin lesions of kits, three different housing systems of does were compared on a commercial farm: single housing, pairs of does, or groups of three does. The aim of the study was to assess these housing systems regarding their effects on rabbit kits. In the following, the results of the study are discussed in detail.

Lesions
As the group size of does increased, the occurrence of lesions in their kits also increased. For rabbits, fights were not observed before 12 weeks of age [5]. For females, Whary et al. [33] suggested that their immaturity is one reason for that. Additionally, food competition should not have been a problem here [1], since there was a sufficient amount of food available and kits were still suckling. Consequently, it is more likely that cage furniture is a potential risk factor for injuries [33]. Nonetheless, in pair and group housing, a much more likely reason is does inflicting injuries on kits. Mykytowycz and Dudzinski [12] observed that does tended to be aggressive towards kits from litters other than their own, with a strong tendency for aggressiveness to increase with the age of kits. In the wild, this rarely occurs, since a series of territorial arrangements protects the kits from being attacked by unfamiliar does [12]. According to Buijs et al. [7], does are also more aggressive in close proximity to their nests. Thus, due to space limitations in fattening cages, the likelihood of social stress and lesions may increase. On the contrary, Rommers et al. [10] observed that in group housing, does often suckled a mix of their own and other kits. However, at the age of 22-30 days pp, when the kits are weaned and the time of birth of the next litter is approaching, the does also become more aggressive

Discussion
In this pilot study, HCC of rabbit kits were presented as a possible new marker for the welfare of young animals from birth to weaning. Based on HCC and skin lesions of kits, three different housing systems of does were compared on a commercial farm: single housing, pairs of does, or groups of three does. The aim of the study was to assess these housing systems regarding their effects on rabbit kits. In the following, the results of the study are discussed in detail.

Lesions
As the group size of does increased, the occurrence of lesions in their kits also increased. For rabbits, fights were not observed before 12 weeks of age [5]. For females, Whary et al. [33] suggested that their immaturity is one reason for that. Additionally, food competition should not have been a problem here [1], since there was a sufficient amount of food available and kits were still suckling. Consequently, it is more likely that cage furniture is a potential risk factor for injuries [33]. Nonetheless, in pair and group housing, a much more likely reason is does inflicting injuries on kits. Mykytowycz and Dudzinski [12] observed that does tended to be aggressive towards kits from litters other than their own, with a strong tendency for aggressiveness to increase with the age of kits. In the wild, this rarely occurs, since a series of territorial arrangements protects the kits from being attacked by unfamiliar does [12]. According to Buijs et al. [7], does are also more aggressive in close proximity to their nests. Thus, due to space limitations in fattening cages, the likelihood of social stress and lesions may increase. On the contrary, Rommers et al. [10] observed that in group housing, does often suckled a mix of their own and other kits. However, at the age of 22-30 days pp, when the kits are weaned and the time of birth of the next litter is approaching, the does also become more aggressive towards their own kits [12]. In future studies, it would also be interesting to investigate whether unfamiliar kits are fully Animals 2023, 13, 196 9 of 12 accepted by the does after cross-fostering or whether the does are more aggressive towards them than towards their blood-related kits.

Hair Corticosterone Concentration
Since rabbits are born naked [34], the corticosterone concentration in the clipped hair is assumed to resample the long-term stress experienced since their birth and is comparable with the shave/re-shave method [15,20]. However, it is quite possible that the period from birth is not fully reflected when HCC is measured in the kits' hair, since there is hardly any scientific data on the speed at which hair grows in kits. According to Perdue et al. [35], by 8 days of age, hair growth in kits is considerable. It can therefore be assumed that hair formation begins well before the 8th day. Interestingly, the occurrence of lesions or disease symptoms in this study was not reflected by the HCC of kits. One reason for this could be that the health status diagnosis is a snapshot compared to the HCC which should be an indicator for long-term stress. Consequently, the HCC value might be lower than expected for some kits because they might have been recently injured, or higher than expected because some kits might have been sick or injured and recovered by the time of the health status assessment. The mechanisms concerning the incorporation of the hormone in the hair shaft have not been fully investigated yet. Nevertheless, it is currently assumed that hair GCs reflect adrenocortical activity [36], and hair has become a favored substrate for GC concentration analysis [37]. Consequently, hair GC concentration analysis has already been used for more than 40 species [38]. However, to the best of our knowledge, it has not yet been used for rabbit kits. This gap should be closed with the present study, in which HCC values for kits are now presented.
A significant connection between the group size of the does and their kits' HCC was not found. However, there is still the possibility that a connection exists, like the tendency for higher HCC values in groups of three does compared to those in pairs of does implies. In a prior study of our research group, there was also no clear tendency when comparing the hair cortisol concentration of kits from does housed singly or in groups of three or five [39]. Trocino et al. [40] found that fattening rabbits of a HYPLUS crossbred line showed higher HCC values in collective pens (20-54 rabbits/pen) compared to pair housing. Nonetheless, these findings might not be applicable to our study due to the large gap between the group sizes and the age of the rabbits. Again, in other mammals, namely sheep and meerkats, the salivary and fecal GC concentrations decreased with increasing group size [41,42].
A significant influence of sex on HCC was also not seen in our study. Cabezas et al. [43] found no significant differences in serum corticosterone and fecal GC metabolite levels enhanced by stress between male and female wild rabbits, either. On the other hand, Monclús et al. [44] observed that male wild rabbits experienced a higher increase in corticosterone metabolites in feces than females due to fox odor. It is possible that the GC response of males and females depends on the type of stressor and the age of the rabbits. Considering other species where the hair GC concentration was measured, in polar bears [17,45] and black bears [46], as well as in non-human primates [47], an influence of sex was found, but not in wild house mice [48], wild white-footed mice [49], rats [50], and in grizzly bears [51]. However, it must be taken into account that the animals in all these studies were already sexually mature, which could also be an explanation for why there was no difference between male and female kits in our study. For example, Waterhouse et al. [52] stated that higher corticosterone levels in female pikas compared to male ones most likely reflected the physiological demands of reproduction.
There also might be other reasons for increased HCC values than those of group size of the does and the sex of the kits. Janicki et al. [53] suggested that cage breeding induces a certain amount of constant stress for adult brown hares, which is; however, insufficient to be considered a significant influence on reproduction efficiency and health status. Morgan and Tromborg [54] listed many other stressors for animals in captivity that can lead to increased GC production. Consequences can be, for example, increased aggressiveness and self-injury [54]. For snowshoe hares, maternal stress during parturition also alters HPA activity and GC levels in offspring, leading to reduced growth and viability of the kits [55]. These effects are also multigenerational, with persistent increases in HPA activity and reduced fecundity also affecting the litters of offspring born to stressed females [45,56]. This inheritance of stress could also be a reason caged rabbits may be permanently stressed, which may have masked the effects of the different housing types on HCC in our study. The reasons for stress in does should therefore also be investigated and minimized.

Conclusions
In this pilot study, the effects of keeping does in groups were investigated; however the focus was not on the does, as has been usual in previous studies, but on the welfare of their kits. Furthermore, HCC values in rabbit kits were presented for the first time. It was shown that the treatment "group housing of three does" had the most kits with lesions and high HCC values. Statistically significant results were only found for lesions. In contrast, it was also shown that pair housing of does did not seem to be disadvantageous compared to single housing of does when considering welfare indicators for kits. From this perspective, a group size of two does seems to have the potential to facilitate social interaction between does without unduly affecting the welfare of the kits. However, this should be considered a preliminary result due to the limited sample size of this study.
In view of this pilot study, further research should be carried out that takes into account not only the welfare of does but also that of their kits in order to evaluate these initial results using a larger sample size. Furthermore, a comparison of HCC of does and their kits would be interesting for future studies.  Institutional Review Board Statement: In compliance with European Directive 2010/63/EC Article 1(5.) f., the present study did not employ any invasive procedure or treatment of the animals. The authors confirm that the study was conducted in accordance with current German law. This study was reviewed and received approval from the Animal Welfare Officer of the University of Veterinary Medicine Hannover, Foundation, Hannover, Germany (TVO-2017-V-83).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on reasonable request from the corresponding author. The data are not publicly available due to privacy.